Enabling data science at DairyNZ with modern science workflows

The New Zealand dairy sector is experiencing a rapid expansion in the volume, variety, and velocity of data generated through research and analysis. This includes data from cow wearables, methane sensors, forage sampling, and satellite imagery. Traditional workflows are often reliant on spreadsheet-based tools which are no longer sufficient to manage, analyse, and share this data effectively. In response, DairyNZ launched the Modern Science Workflows (MSW) initiative to build scalable infrastructure, enhance data science capability, and embed reproducible business disciplines across the organisation.The MSW project was structured around three pillars: infrastructure, capability development, and business disciplines. Infrastructure upgrades included the adoption of Snowflake, Posit Workbench and Posit Connect. These tools enabled researchers to work with datasets exceeding beyond the limits of traditional Excel workflows. Capability development was delivered through structured R for Data Science courses, complemented by internal statistics training and regular R community meetings. Business disciplines were reinforced through GitHub adoption for version control, documentation standards, and alignment with organisational data governance.The initiative has already delivered measurable outcomes. Projects using high-resolution animal sensor data such as predicting grazing performance and heat stress have been enabled by the new infrastructure for both data storage, modelling and deployment. The Economics team has championed the use of R, RStudio, GitHub, and Posit tools in sector-wide outputs including the DairyNZ Economic Survey and Econ Tracker, published via Posit Connect. Survey feedback indicates strong uptake and appreciation of the training programs, and the organisation is now transitioning toward an open science model. Plans are underway to deploy an organisational GitHub account and cataloguing tool to improve code discoverability, ownership, and collaboration.ABOUT THE AUTHORBen Marmont is an Economist at DairyNZ where he has been involved in the Modern Science Workflows project for four years. During this time, he has been involved in; capability development by delivering R for Data Science courses as both a tutor and lecturer alongside chairing the R Community, business disciplines by facilitating the deployment of the organisational GitHub, and infrastructure by being internal early adopter. -----For more information about the eResearch NZ / eRangahau Aotearoa conference, visit: https://eresearchnz.co.nz/